From a previous question I gained an understanding of how intersection works among multiple domains and ranges of a predicate.
It seems that union is the more common way to handle this scenario, so now I'm wondering what different inferences are made from a union of domains or ranges.
Given p rdfs:domain (C ∪ D ∪ E) and the triple a p b, what is inferred about a?
For those who are not well versed in Set theory (like myself) I will provide some explanation of what I think it means for a to belong to the union of C, D, and E.
a is at least one of C or D or E.a may be more than one of C and D and E.
a may be type C and type E (or may be all three types).C, D, and E all have a common superclass S, then a is also of type S.a beyond these "maybes". Specifically, we cannot infer that a has more than one type, but we cannot infer that a has less than three types either.